BIOMARKERS IN THE PREDICTION OF METASTASIS IN MELANOMA

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

Melanoma is a potentially lethal malignancy that arises in the skin. Its incidence and mortality rates continue to increase. Recently, new staging procedures and adjuvant therapies have been developed that may improve survival. However, these are variably expensive, intrusive, and toxic. Accurate prediction of the probability of metastasis in patents presenting for therapy of primary melanoma will allow for improved management and outcome. It will protect patients with disease without metastatic capacity from unnecessary and costly interventions. For patients with melanomas that may metastatic capacity from unnecessary and costly interventions. For patients with melanomas that may metastasize, it will allow for improved clinical decision making by better matching risk of metastasis to the risks and costs of staging and adjuvant therapies. The objective of this project is to develop a credible, accurate and generalizable prognostic model that will be effective in managing patients. Explicit goals are accurately to dichotomize patients into those with disease incapable of metastasis and those with more advanced disease and to predict the latter's probability of having metastases. Primary melanomas evolve in steps and this process of tumor progression provides the framework for prognostic modeling and related clinical decision making. A critical event in tumor progression in primary melanoma is the transition from an invasive lesion without metastatic ability to one with competence for metastasis. In this project we will use clinical and pathological data from a cohort of patients followed for 10 years as variables in the construction and validation of: 1) prediction rule to identify non-metastasizing lesions and 2) a probability model for risk of metastasis in patients with more advanced primaries. Candidate variables will include the reactivity of monoclonal antibodies (MAbs) to molecular markers of melanoma cell proliferation and the phenotype and lytic capacity of tumor infiltrating lymphocytes. To assure that the model is generalizable, expression of these molecules will be detected in routinely fixed and paraffin-embedded material with available MAbs and standard immunohistologic techniques. These variables will also be studied in a prospective series of patients to test as an independent variable a peripheral blood assay for expression of the gene for of the pigment-related protein tyrosinase.